A cost-minimisation analysis of performing point-of-care ultrasonography on patients with vaginal bleeding in early pregnancy in general practice: a decision analytical model

Abstract Background Spotting and light vaginal bleeding are common and usually harmless symptoms in early pregnancy. Still, vaginal bleeding may be the first sign of an abortion and often causes distress to pregnant women and leads to an expectation of an ultrasonography examination of the uterus. As point-of-care ultrasonography (POCUS) is increasingly being integrated into general practice, these patients may be clinically evaluated and managed by general practitioners (GPs). This can potentially reduce referrals of patients from the primary to the secondary healthcare sector resulting in societal cost-savings. The primary purpose of this study was to investigate whether the accessibility of POCUS in general practice for patients with vaginal bleeding in early pregnancy is cost-saving compared to usual practice where GPs do not have access to POCUS. A secondary purpose of this study was to estimate a remuneration for GPs performing POCUS on these patients in general practice. Methods A cost-minimisation analysis was based on a decision tree model reflecting the two alternatives: general practice with and without GPs having access to POCUS. The robustness of the model results was investigated using probabilistic sensitivity analysis and the following deterministic sensitivity analyses: one-way analyses for the model input parameters and a scenario analysis with a change from a societal to a healthcare sector perspective. An expected remuneration reflecting the add-on cost of Danish GPs performing POCUS was estimated based on the related costs: cost of an ultrasonography scanner, GP’s time consumption, ultrasonography training, and utensils per scanning. Results The difference in average cost between the two alternatives from a societal perspective was estimated to be €110, in favour of general practice with GPs using POCUS. The deterministic sensitivity analyses demonstrated robustness of the results to plausible changes in the input parameters. The expected remuneration for performing POCUS in this specific setting was estimated to be €32 per examination. Conclusion Having GPs perform POCUS on patients with vaginal bleeding in early pregnancy is cost-saving compared to usual practice. The results should be taken with caution as this study was based on early modelling with uncertainties associated with the input parameters in the model.